Steve Ballmer: Stanford Graduate School of Business

Remarks by Steve Ballmer, Microsoft Chief Executive Office
Stanford University – Stanford, Calif.
Sept. 25, 2008

Editor’s note, Sept. 26, 2008 –
Due to a technical problem recording the event, the first part of Steve Ballmer’s remarks aren’t captured in the transcript below.

STEVE BALLMER: (In progress) — little thing going. Even our board says, you really have 30 priorities in there. Of course, we don’t have 30 priorities, but there are 30 key indicators of how we’re doing, and we’d better have somebody in the company for whom every one of those 30 things is a priority.

It’s part of the transition you make when a company goes from being really in one business to being in multiple businesses, and it just means that if you’re going to perform well, you’ve got to perform well more broadly.

So, yeah, we’ve got 30 things we look at. The scorecard forms the basis for my review. Some of the things are innovation-based, some of them are employee development and excellence based, some of them are innovation based, some of them are kind of marketing and image based, but those would be kind of the big buckets.

MODERATOR: You spoke a bit to the evolution that the company has gone through over many years, and you’ve been there since very early obviously. How have the leadership challenges evolved for you since when you got there in the early ’80s?

STEVE BALLMER: In a way it’s almost like two completely different universes. What it takes to be a good leader of a 30, 40, 50, 100-person organization is different than what it takes to be a good leader of a thousand, 2,000-person organization in a single business, is different again than what it takes to be the leader of a — we’re actually closer to 90,000 employees now — person company that’s in numbers of businesses. What you do, how you spend your time, how you think about things is completely different.

I mean, there was a day when I would tell you — when I would have said, I run the business, I feel like I — I don’t run our business. I hope the guys who are running our business are running our business — (laughter) — and I’d better hold them accountable for it.

But running the business sort of says every day you’ve got your hands, you’re moving levers. That’s not what I’m doing. I’m picking people, I’m picking strategies, I’m spending my time — I’m doing a lot of external representation.

But day-to-day who’s going to make — there’s a few big calls that would pop out, but who’s running every day our enterprise business? It’s not me. And yet I would have told you, even as recently as four or five years ago, I would have felt like I had my hands on many more of our businesses. But when you get as diverse, you’ve got to recraft and rethink your role.

So, today I don’t feel like I operate. When you’re 30 people, man, that’s all you do is operate, and then there’s kind of a variety of phases in between.

MODERATOR: So, what was the toughest part of that transition for you, going from the hands-on operator to your current role?

STEVE BALLMER: My natural personality is more the hands-on operator than — I just love, you know, give me details, give me things to handle, handle, handle. I mean, rrrrrrr. (Laughter.) And it’s different. It doesn’t make it better or worse; it makes it different.

So, really I think if you talk to folks, they wouldn’t exactly use the word micromanager, but I like detail. (Laughter.) That math tournament thing, comes back to play.

So, really the question is — and I can’t even pretend – is our business is diverse enough. I can’t be — I can’t know enough in depth to really try to figure out what we should do if I don’t agree with the guys who are running the business. I have to raise questions. I’ve got to hold people accountable. There are big parts of our businesses that have been around longer that I know better. I mean, I know the Windows business very well. I have a better opportunity to really provide input and change. But, you know, look, I know the advertising business, but I’ve never operated our advertising business, and it just puts you in a different role. That’s not a mistake. You can’t have a single human being who — I mean, we can’t do it on the technical side either. Ray Ozzie needs to know what links our business technologically, but will he really be in depth in every one of the businesses we’re in? It’s just nonsensical. You’ve got to really delegate and trust people.

MODERATOR: So, speaking of those people that you trust to be at the controls every day, what processes do you have in place to retain those employees, outside of compensation obviously. But they’re clearly targets every day to be picked off by other technology companies or move down to Silicon Valley, whatever it may be. What are you doing to keep those people in play?

STEVE BALLMER: Well, keep housing prices down; high up here, it helps. (Laughter.) But no, I’m teasing.

You know, truthfully we get very little, I’ll call it poaching, of our senior-most people; extremely low levels of poaching of our top thousand or so people, what we call our partners. It doesn’t mean it doesn’t happen occasionally, but very low. These are generally people who have been with us a long time, have a lot of loyalty, really believe in the business. Occasionally, somebody will want to leave, different kind of opportunity, and usually they’ll talk to me a long time in advance. We had a fellow leave recently to become CEO of Juniper. He had talked to me for a year or more about, hey, I might really want to be a CEO some day, and I told him why I didn’t think that was all that exciting — (laughter) — but, hey, eventually that was what he wanted to do, and you move forward.

Frankly, what I would call the retention issues are far more dramatic kind of in the next band of people, the people who are sort of up-and-comers, maybe, what would I say, four years to 10 years into their career. That’s where we have sort of the most work we have to do, and you’ve got to make sure you’re giving people opportunity as fast as they can take it. You’ve got to make sure you’re giving people rewards and compensation as fast as they can take it.

At the senior levels, the real issues are how do you mold things so that people feel like they have enough individual autonomy and authority, and yet the whole is bigger than the sum of the parts. I’d say that’s the number one challenge in what I might call that top thousand type group.

MODERATOR: And how much of your time are you spending developing those people, those top thousand, focusing on them, figuring out their next role? Realistically it’s probably not a thousand folks that you’re daily dealing with, but how much of your time goes to that?

STEVE BALLMER: I’d say there’s probably about 300 or 400 people I really feel like I should have — can have and should have some real involvement in helping shape what they do next. That happens in a variety of ways. Once a year we do a big day or two-days per business where all we do is talk about the strategy, and then we talk about the key people and how we apply the key people to the strategy and how that shapes going forward; big exercise.

I do one-on-ones with my directs roughly monthly, depending on the business.

It turns out that all the time in one-on-ones or most of the time in one-on-ones is actually about people, because most of the time that you want to talk about strategy. You actually want to have a broader set of people than not, and so you don’t tend to do the strategy discussions as much in the one-on-one, you tend to do people stuff. So, you could say there’s another 12 times 10 – 120 hours or so a year – where we talk about people.

I was just at an offsite. We do four offsites a year with our Senior Leadership Team. We spent probably four hours talking through people as a team.

So, I would say it’s a lot — we have a review process. Anybody who’s going to get promoted beyond a certain level we talk about the people as a Senior Leadership Team.

So, it is a huge part of picking, choosing, developing. We do what we call a strategy conference three times a year where we bring people who are up and coming in various different capacities in their careers, we bring them together. We spend a day and a half getting to know them, talking about what’s going on, hearing their views on strategy.

So, nurturing, developing the top talent I would say probably for me it winds up being the number one thing I spend time on.

MODERATOR: Switching gears a bit to a very recent public event for Microsoft, what key capabilities were you hoping to bring in-house with the Yahoo! deal, and how do you plan to build those capabilities now that the deal isn’t going through?

STEVE BALLMER: Twenty percent search share. (Laughter.) No, I’m actually dead serious. I deadpanned it for fun.

But the truth of the matter is we’ve got very good people, Yahoo! has had good people; that’s not really the issue.

But you don’t spend US$40 billion or whatever the amount would have wound up being, you don’t spend that to buy capability. You actually spend it to buy position in the marketplace and with customers.

The search advertising market actually is a very unusual market, because it is an auction marketplace, and because the advertising isn’t an add-on to the product, it’s a fundamental part of the product.

So, when you look at a search page from us or from Google or Yahoo!, there’s probably three things that you use to evaluate the quality of the results. One is what we pump out algorithmically and generally put on the 10 blue links, as we call them, on the lower left side. We have absolutely great plans to catch, surpass, et cetera. The user interface probably needs a revamp; it never changes. Because we’re not the market leader, it’s probably a little easier for us to play with that than it is the market leader.

The second thing you evaluate is the relevance and quality of the advertising. Well, it turns out that the more — if you want to have more relevant ads, you actually have to have more advertisers bidding on more keywords. The more advertisers that bid, the more keywords they bid on, the more you can pick and choose exactly the right ad. But the ads are part of that user experience.

Because we have 10 percent and Yahoo! has 20 percent and Google has whatever — and this is in the U.S.; outside the U.S. Google has more, 60-odd percent — what you get is a world in which there’s more bidders bidding on more keywords in the Google system. This says nothing about technology, capability, blah, blah, blah; it’s just the fact of the matter is you deliver better results, and, by the way, because they have more bidders on more keywords, they also have a chance to make more money.

So, it’s both a product quality issue and a monetization opportunity if you can get more critical mass of users, so you get more critical mass of advertisers.

So, when you ask me what were we trying to get, we were trying to get 20 percent — no, there are other great things in Yahoo!, but number one is 20 percent search share in the United States.

The third part, frankly, of the search page is the brand. You show the same results absolutely side by side, and you put the brand of the market leader on it, people are going to rate the same results higher. You would expect that; that proves marketing and brand means something. People rate them higher.

So, we have some work to do on brand that probably was not solved by the Yahoo! acquisition. But in terms of monetization and ad relevance, getting scale Microsoft and scale Yahoo! together was kind of at the core I would say of that strategy.

MODERATOR: So, assuming that that deal can’t be revived, what do you do as Microsoft — I think it would be difficult probably to merge with Google. (Laughter.) So, what are your choices?

STEVE BALLMER: Neither one of us are antitrust attorneys, but I’ll bet you’re right. (Laughter.)

MODERATOR: You heard it here.

STEVE BALLMER: Well, there are two things. Of course, first we went around where we actually tried to see if there wasn’t a way to collaborate on search, even if we weren’t acquiring Yahoo! We worked on that with Yahoo!, we worked on that with Carl Icahn. We worked on that. That didn’t work. (Laughter.) Eh, it’s okay. They know where to find us if they’re ever interested.

So, what we do now is we have to go build through marketing, through sales, through promotional activities, distribution activities, we have to go build ourselves kind of our organic base. It will take longer. I’m not sure it will be more expensive to our shareholders. Well, I mean, you say you’re going to spend $50 billion, that’s a lot of money. It’s ironic, $50 billion on the balance sheet sometimes looks cheaper to shareholders than a billion dollars a year for five years. Of course, a billion dollars for five years is 5 billion, and 50 is 50. Yeah, there’s a profit stream that comes with the 50, but people tend to apples and oranges comparison because the accounting rules actually make acquisitions always look favorable from an investor perspective. It’s not smart, but it’s the way most investors look at things as opposed to in real economic terms.

So, in any event, we’ll have to take more through the operating income, a hit more on operating income as opposed to a balance sheet hit, and it will probably take us a little bit longer. I did tell investors that they should be prepared for us to invest someplace between 5 and 10 percent of contribution margin back into our search business every year for the next several years, and that’s kind of the frame for where we go from here.

And we’ve got a lot of innovation and sort of things we need to do to change the experience, change the business model. When you’re not the market leader — my most important lesson from first year at Stanford Business School, strategy course, Professor Wheelwright, and I can’t remember what the heck the case was, the market leader should increase fixed costs, the number two person in the market is the one who wants to either cut price or raise variable costs. We’re kind of clearly not number one, so we clearly want to play the game on more of a variable cost and pricing basis than on a fixed cost basis.

MODERATOR: Are there any decisions you made over the course of your career that might have seemed minor or trivial at the time, but have had a significant impact on your evolution as a leader in retrospect?

STEVE BALLMER: As a leader.

MODERATOR: More on the company.

STEVE BALLMER: Well, on the company the answer is for sure. I mean, still the most important acquisition the company has ever done was the purchase of a company called Dynamical Systems, which I think we probably paid, I don’t know, $7 million for or something like that. There was no venture money. It was a bunch of PhD physicists from Princeton, and their buddies. They were programming out of a house in Berkeley. We bought the company. They had something that was a clone of a weird, kooky product that IBM had that we thought we needed in order to sell IBM something, blah, blah, blah. None of that ever mattered, but we got some of the guys who provided some of the core architectural and engineering that really made Windows a success.

So, you say it seemed like A, it turned into B, but the guy basically who really helped put Windows on the map in a technology sense is the guy who came through that acquisition. The CEO of that company wound up being our kind of CTO for a lot of years.

So, yeah, I think we’ve had some of those accidental things that wound up being quite profound.

MODERATOR: And then how about on you as a leader? Anything?

STEVE BALLMER: On me as a leader. Things that were small at the time that then felt bigger. Harder for me to point to that kind of case. I think the kinds of things that I’m drawn to are people who I may not have really grokked or valued or appreciated when I first met them. They might have seemed quirky or not so capable in one way or another, and they wound up being some of the folks who made the most sort of profound difference on the place.

One guy in particular I think of. He’s not himself the amazing story. It was a guy who every time I met him, I thought, oh geez, is this guy really going to make it at the company? I’ll say we’re kind of an IQ-oriented company, and I’d say, God, does this guy have a high enough IQ to do his job; I don’t know. Is the guy going to be able to get there? The guy taught me more about what leadership and management is really about, because he never really had the whole picture and understood all the details, but he really knew how to build a team.

Sort of great leaders are part thought leaders, part business leaders, and part great people pickers and people managers. I think most people tend to sort of focus more on any one of the three of those, and I was certainly more on thought leadership, it’s all about thought leadership. Some people tend it’s all about your ability to manage people. But the truth is great leaders have to have a mix of those things, and I think in a way I got taught that by this one country manager who I kept thinking was no good, and then the business just exploded, his people wound up running various big parts of the company globally, and I learned really to appreciate a style that was quite different perhaps than my own, and a set of capabilities different than my own.

MODERATOR: And did that lead to any specific changes at Microsoft in the way you thought about performance management or evaluation of some of the key people there?

STEVE BALLMER: I think it made me and perhaps others far more open to the notion that there were sort of a variety of different approaches to leadership that could be quite successful, where I would say our classic model is one person, everything in their head, go figure it out.

Well, I described it doesn’t work in my current job, but I will say it worked fine when we were 30 people, worked fine when we were 100 people, a thousand people, 3,000 people, and maybe it was two guys heads, mine and Bill’s, but the system breaks down at some point, because at some point you’re not operating one business, and then you’re really got to appreciate these kinds of skill sets.

So, yeah, I think it actually did make a difference, because it made us all more open. It doesn’t mean you can go all with people of that ilk either. You’ve got to have some leaders who are primarily thought leaders. But mixing it up and getting the right balance in the team, yeah, I think it came out of that experience.

MODERATOR: If you could change one decision that you’ve made since becoming CEO, what would that be?

STEVE BALLMER: Since becoming CEO. So, I’ve got to get the timing right. Well, it was a non-decision. We should have started our R&D efforts in search earlier; should have, no question. No question. And I don’t say we should have bought Inktomi or we should have bought Overture; we should have bought something, maybe we should have done that, too, but we should have started. We really got in the R&D business around search, I can’t remember, four years ago now, five years ago. If we had gotten in eight years ago, I think we’d be in a very different place than we are today.

MODERATOR: And was there a significant discussion of that opportunity eight years ago, or was it just not on the radar screen the way other things were?

STEVE BALLMER: Some discussion, but this is one of the challenges that sort of relates to the kinds of challenges that you were asking about at the beginning. When you’ve got your hands full and a new great opportunity comes along, it takes a lot of, what shall I say, discipline to do both and to do both well.

One of the things organizations can do is discount the size of the opportunity, which was not unfair at the time. I mean, search has been around for a long time. In some senses, the great innovation of Google isn’t that there’s a thing called Search. AltaVista, there are people who did that stuff before. The great innovation was, hmm, there’s a magic way of using a software-based auction marketplace to make money out of the thing.

I’m not sure we even needed to have that insight at the time. We still should have gotten started, even though the business model wasn’t clear, because at the end of the day things that users value, you know, necessity, the mother of invention, you’ll figure out how to monetize them if they bring end users value.

MODERATOR: And have you made any structural changes in your organization to at least try to prevent missing another opportunity like that or at least being late to the game?

STEVE BALLMER: Yeah, we’ve made a lot of changes actually. We have more of what we call these lab type groups, which are kind of incubation. The current fashionable thing at Microsoft is to take your product name and have product name labs: Office, Office Labs; Live, Live Labs, you know, Virtualization, Virtualization Labs. It doesn’t matter. And labs says, I’m incubating, I’m doing some stuff with small teams to try to make sure we’re planting more trees. That’s certainly taking shape.

The way we run our annual sort of business strategy process has changed as a result of that. We have kind of an outside-in where is money being created, not just what are we doing or what are the strategies for the businesses that we’re in. Our research group we just continue to scale up, because our research group was kind of all over this, and kind of — I mean, when we finally got in the game, thank God we had a research group, because there was all kinds of talent in information retrieval and search and the like.

So, we’ve pushed the pedal down further. But at the end of the day, we have to be prepared both to lead and to fast follow, because if you’re as big a footprint as we are, it would be great to say we’ll invent everything that’s ever going to be important — that’s probably not practical. Let’s try to invent the highest percent we can, but on other things that are going to be really important, it’s important also to be able to be a good fast follower. That’s not a strategy, but it is a muscle you need to have, too.

And if you take a look at it, a lot of the things that we all go goo-goo, gaga over now, they weren’t original concepts invented by the companies who made them popular, but they were follow-on efforts where somebody was a fast or a better follower. You want to have both muscles available inside our company or maybe any company of any size.

MODERATOR: Microsoft is at the center of significant media attention, both positive and negative. What steps are you taking to lead through that noise with your organization?

STEVE BALLMER: How do I say this the right way? (Laughter.) We’ve had so much noise now for so many years, it’s easier to lead through the noise today than it was — the height of noisiness and issue with our employee base on this would have been the height of our antitrust issues really with the U.S. Department of Justice, because that’s our home country where we have sort of the largest percentage of our employees. And it was not easy, because when your own government sues you, people assume that there was something wrong. I mean that in kind of not just a — you know, nobody really knows the law of large market share. Once you have it, the government does its evaluation, decides if it’s too long, what are the rules; that’s the way the world works.

But people say, why didn’t they know? They should have known. How could they not know? Well, you can’t know, and so you kind of have to reassure employees about the integrity of the institution, the people, the fact that we were doing the right things all along in a sense, and fine, there’s a certain set of rules and we’ve got to follow those.

So, you kind of have to do a lot of heavy lifting, I would say, at that time. People do a lot of soul-searching. Our leadership team went offsite, I don’t remember what year, probably ’01, ’02, something like that. We went for an offsite, and we thought we were going to have a normal offsite, and we wound up talking just about our values, and we wound up writing down the values of the place, because it was so important at that time to communicate internally with our employees, externally, et cetera.

You could say every company should do it at some point, but in our case I would say it was brought on — we’d always been busy, and I think we’d be high values, but you want to write them down and let people know what you’re thinking.

Now the thing is, you know, up, down, you’ve still got to talk to your employees. The number one thing I would say our employees have on their minds of that ilk is when our products are good they know, when our products aren’t as good they know. So, we don’t have to talk to our employees as much about that. We have to say what we are doing. But we have to talk to our employees these days a little bit about the stock price.

Bill and I had always vowed we would never talk about the stock price, because as Graham said, in the short term it’s a voting machine, in the long term the stock price is a good weighing machine; so any day you look at it. But employees look at it like a day-to-day kind of report card, how did we do today. Oh, we’re up three percent today. Oh, we did brilliant work today. (Laughter.) Of course, it’s not how it works, but our stock has been flat for a while, despite the fact essentially if you add back dividends, we’re largely flat for 10 years. People say, why would — and that’s not actually untrue of a lot of other large cap stocks. A lot of large cap stocks have that profile. But the question is why – are we doing the right stuff? At the end of the day if you drive earnings, you are doing the right stuff. And in our case though the market is also voting on today’s products, tomorrow’s opportunity, today’s capital structure versus what they might want for a capital structure. So, really educating employees so that they can make their own value sort of estimations and determinations about how the company is doing as opposed to trying to look at a very ineffective day-to-day scorecard, we spend a little bit of time on that I would say.

MODERATOR: That process of writing down and communicating the company’s values, which were always there, but you had an opportunity to explain to the employees a little bit more clearly, did that make a difference? Did it help a bit as it played into your leadership?

STEVE BALLMER: It helped certainly kind of give people a magnetic north at the time. It helped remind people. We baked it, we put it in our review system for three, four years, and people got reviewed on the values, in addition to their performance. It turns out that’s sort of a hard one to do. Integrity: You sort of either get a you’re fired or you’re okay. (Laughter.) There is kind of a binary on some of this stuff. But people understood our seriousness about those things as well, and I think that was helpful.

Now, it got to be kind of redundant doing it every year, so we focused back now primarily on sort of normal performance, but I still talk all the time, I’ll quote from various of our values often in the kinds of sort of speeches and broad communications to employees.

MODERATOR: You’ve interacted with many CEOs, not only in the software space but also in other industries. Is there a leader or two whom you most admire, and why?

STEVE BALLMER: I’m sort of not an admiring kind of a guy. (Laughter.) No, I don’t mean that in a — admire sort of implies a blind and whole acceptance, and that’s not my personality. I can tell you things that I learn, things that I respect, and where I say, hmm, that person really has a good sense of that. Boy, really it was great to hear Jeff Immelt. Jeff and I played golf together a couple of years ago, and he was talking about some things he had to do with one of his guys. He said, I’m going to go see him to see whether he still has his confidence, because if he doesn’t have his confidence, I’m going to have to take him out of his job. It’s a good point. When people lose their confidence, you can’t lead big organizations without your own self-confidence, you can’t.

You learn something there, you learn — you know, we were talking about A. G. Lafley. Just listen to A. G. Lafley talk about spending time in China watching women do laundry. You say, what do I care about learning laundry? That’s a different kind of guy. That’s a guy whose consumer research and sort of fundamental push to understand, really to emphasize, really, and whether it’s China or Germany or U.S., that’s kind of where they live. And he’s got to set a tone for everybody: I expect you, who runs this business, to be out there and know these kinds of customers as well as I know these kinds of customers.

I think these are all great people, but I wouldn’t say I admire them. I say, God, I learned something there; God, that guy is good at this; God, this woman is good at this; what about this, what about this, what about this. Every interaction basically, not just with CEOs, but every interaction that I have with talented people I think you get a chance to glean. My job – I get a chance to have more interactions with more talented people than most, and I feel fortunate that I’m able to sort of glean those kinds of practices from a variety — Mukesh Ambani. Mukesh was my class at Stanford. He and I both dropped out, and he went to help his dad with the family textile business, and it’s now, between the company he runs and his brother runs, it’s like 40 percent of the Indian stock market.

You go back and see Mukesh in India, and for a while we were talking to him about a telco project, and he said, the price of the service is going to be this, so you have to price your stuff at this, otherwise we can’t do it. And I say, well, we don’t know how to do that, and he said, well, you’d better go figure it out, because I know exactly what the Indian consumer, mass Indian consumer can afford to pay.

Today, he’s building as much retail space as like all of Wal-Mart just in India. He says, look, I learned from my dad. If you want to sell to the lower classes in India, you’ve got to let them make money from you. So, he’s got a whole system of buying fruits and vegetables from farmers in rural India so he can turn around and bring them into the retail stores that he’s building in those parts. I mean, real kind of savvy consumer understanding, you learn something from a variety of people in a variety of different ways, and I think you just want to get every nugget you can.

MODERATOR: We’re going to open up the floor for questions, but I have one other one for you, which is what would some of those key leaders say about you, say what they’ve learned about your leadership style or how it reflects on them?

STEVE BALLMER: Yeah, I think probably the two things people would tend to comment on — this is kind of a task in self-awareness I guess — (laughter) — with three cameras flying at you, but I think — (laughter) — I think a couple things. I think that I certainly have more time invested thinking about how you run an innovation-based company than most people do. I mean, the truth is we make big bets, they are long-term. We’ll spend, whatever, 9-odd billion, something like that, in R&D. So, we do things at kind of a scale. We have big projects, little projects. We’ve got incubations and research, we’ve got acquisitions and internal development.

I certainly have talked about that a number of times with a number of CEOs, and I think people at least understand that whether I think smarter or not about it, I certainly think about it a lot.

And then I think the second thing, which my YouTube videos really get a lot of CEO attention, too. (Laughter.) And it turns out a lot of guys will say, do you really do that? There is something where people say, hmm, how do you think about your presentation style and what you’re trying to communicate to your people? I mean, I’m a kind of rah-rah passionate blah-blah enthusiasm kind of guy, and some CEOs will say, boy that would never work at our place. But at least they’re kind of in the thinking loop about what does it take for any given population at Microsoft or anybody else’s to keep the kind of employee base focused, enthusiastic, excited, et cetera.

MODERATOR: Very good, thank you.